gg-tt (gg-tt)

posted an update 3 days ago

Post

3632

We’re releasing new Qwen3.6 quants that run 2.5× faster on your GPU. ⚡

Qwen3.6-27B NVFP4 runs on 24GB VRAM.
35B-A3B can hit 17,561 tok/s (B200).

We also improved accuracy, tool calling, agent use, and looping.

Qwen3.6 NVFP4: https://huggingface.co/collections/unsloth/nvfp4
Guide: https://unsloth.ai/docs/models/qwen3.6#nvfp4

1 reply

·

osanseviero

authored a paper 4 days ago

Gemma 4 Technical Report

Paper • 2607.02770 • Published 11 days ago • 60

osanseviero

submitted a paper to Daily Papers 5 days ago

Gemma 4 Technical Report

Paper • 2607.02770 • Published 11 days ago • 60

danielhanchen

posted an update 6 days ago

Post

5786

DeepSeek-V4 can now run locally with Unsloth GGUFs! 🐳

Run lossless DeepSeek-V4-Flash on 168GB RAM or
3-bit works on 110GB Mac, RAM, VRAM setups.

Run via Unsloth Studio or llama.cpp.

GGUF: unsloth/DeepSeek-V4-Flash-GGUF
Guide: https://unsloth.ai/docs/models/deepseek-v4

danielhanchen

posted an update 20 days ago

Post

3282

1-bit GLM-5.2 GGUF vs. Claude 4.8 Opus vs. GPT-5.5

We gave 3 models the same prompt and compared one-shot outputs.

The 1-bit GLM-5.2 GGUF ran locally on a Mac Studio M3 Ultra with 256GB RAM at ~21.6 tok/s.

Which output do you like best?
GGUF: unsloth/GLM-5.2-GGUF

3 replies

·

danielhanchen

posted an update 27 days ago

Post

4565

Google's new DiffusionGemma can now run at 2000+ tokens/sec! ⚡

We made local DiffusionGemma inference 1.8× faster.
Run it on 18GB RAM via Unsloth Studio.

GitHub: https://github.com/unslothai/unsloth
Guide: https://unsloth.ai/docs/models/diffusiongemma

4 replies

·

danielhanchen

posted an update about 1 month ago

Post

1161

Google releases DiffusionGemma.✨
The new 26B-A4B diffusion text model runs locally on 18GB RAM.

Run with 4x faster text generation, thinking, image, video and 256K context. Run and train via Unsloth Studio.

GGUF: unsloth/diffusiongemma-26B-A4B-it-GGUF
Guide: https://unsloth.ai/docs/models/diffusiongemma

1 reply

·

danielhanchen

posted an update about 1 month ago

Post

4269

Google releases Gemma 4 QAT. ✨
You can now run Gemma 4 at 3x less memory with near original performance.

QAT makes it possible to run Gemma 4 26B-A4B on 16GB RAM.

GGUFs: https://huggingface.co/collections/unsloth/gemma-4-qat
QAT Guide: https://unsloth.ai/docs/models/gemma-4/qat

1 reply

·

danielhanchen

posted an update about 1 month ago

Post

9296

Gemma 4 12B can now run locally on just 8GB RAM via Dynamic GGUFs.

Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.

GGUF: unsloth/gemma-4-12b-it-GGUF
Guide: https://unsloth.ai/docs/models/gemma-4

5 replies

·

srush

authored a paper about 1 month ago

Composer 2 Technical Report

Paper • 2603.24477 • Published Mar 25 • 20

emozilla

authored a paper about 2 months ago

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

Paper • 2604.27263 • Published May 14 • 11

danielhanchen

posted an update about 2 months ago

Post

2820

Qwen3.6 MTP is here! Run locally on 20GB RAM. ⚡️

MTP enables Qwen3.6 to generate ~1.4–2.2× faster with no accuracy change.

Qwen3.6-27B: unsloth/Qwen3.6-27B-MTP-GGUF
Qwen3.6-35B-A3B: unsloth/Qwen3.6-35B-A3B-MTP-GGUF
Guide: https://unsloth.ai/docs/models/qwen3.6#mtp-guide

2 replies

·

tomaarsen

posted an update about 2 months ago

Post

1225

🤗 Announcing the Ettin Reranker family: six new state-of-the-art CrossEncoder rerankers for search from 17M to 1B parameters, plus the full training data and the ~150-line recipe. Built on the Ettin ModernBERT encoders, Apache 2.0. Details:

All six were trained with the same single-stage pointwise MSE distillation recipe, with mixedbread-ai/mxbai-rerank-large-v2 (1.54B) as the teacher. Only the learning rate and per-device batch size change between sizes. The 1B student matches the teacher within 0.0001 NDCG@10 on MTEB(eng, v2) Retrieval, the 150M is the strongest reranker I tested in the under-600M range, and the 17M beats the 33M ms-marco-MiniLM-L12-v2 by +0.051 NDCG@10 at roughly half the parameter count.

Speed matters as much as quality for a reranker, since it determines whether the model fits the latency budget between retrieval and showing results. Our 17M is the fastest reranker in the whole comparison at 7517 pairs/sec on an H100. Our 150M runs 2.3x faster than the two other 150M ModernBERT-base rerankers (gte-reranker-modernbert-base and granite-embedding-reranker-english-r2) because the modular Transformer module propagates unpadded inputs through every layer rather than just the FA2 attention kernel. And our 1B is 2.4x faster than its 1.5B teacher while matching it on quality.

I bootstrapped the training recipe with the new train-sentence-transformers Agent Skill shipped in Sentence Transformers v5.5.0. Install it with hf skills add train-sentence-transformers --claude and ask Claude Code (or Codex / Cursor / Gemini CLI) to fine-tune a SentenceTransformer, CrossEncoder, or SparseEncoder model on your data.

I wrote a blog post walking through usage, results across six embedder pairings, the speed story, and the complete training script. Check it out, or just point your Agent to the URL:

https://huggingface.co/blog/ettin-reranker

Collection: https://huggingface.co/collections/cross-encoder/ettin-rerankers

emozilla

submitted a paper to Daily Papers about 2 months ago

Targeted Neuron Modulation via Contrastive Pair Search

Paper • 2605.12290 • Published May 12 • 17

emozilla

authored 2 papers about 2 months ago

Long Context Pre-Training with Lighthouse Attention

Paper • 2605.06554 • Published May 7 • 31

Efficient Pre-Training with Token Superposition

Paper • 2605.06546 • Published May 7 • 48

danielhanchen

posted an update 2 months ago

Post

5984

We’re excited to announce that Unsloth has joined the PyTorch Ecosystem! 🔥🦥

Unsloth is an open-source project that makes training & running models more accurate and faster with less compute. Our mission is to make local AI accessible to everyone. Thanks to all of you for making this possible! 💕

Blog: https://unsloth.ai/blog/pytorch
GitHub: https://github.com/unslothai/unsloth

2 replies

·

tomaarsen

posted an update 2 months ago

Post

545

🤖 I've just published Sentence Transformers v5.5.0, headlined by a new train-sentence-transformers Agent Skill that lets your AI coding agent (Claude Code, Codex, Cursor, Gemini CLI, ...) train and finetune embedding, reranker, and sparse encoder models for you. Plus training losses & fixes. Details:

The skill bundles curated guidance for the whole training workflow across all three model types: base model selection, loss and evaluator choice, hard-negative mining, distillation, LoRA, Matryoshka, multilingual training, static embeddings, etc. It also ships production-ready training template scripts the agent can adapt. Install it with hf skills add train-sentence-transformers, then just describe what you want, e.g. "finetune a reranker on my (question, answer) pairs, mine hard negatives, and push it to the Hub".

On the loss side: EmbedDistillLoss is a new embedding-level distillation loss for SentenceTransformer. Instead of distilling teacher scores like MarginMSELoss, it aligns the student's embeddings directly with pre-computed teacher embeddings, wtih an optional learnable projection for when the student and teacher dimensions differ. Second, ADRMSELoss is a new listwise learning-to-rank loss for CrossEncoder from the Rank-DistilLLM paper, aimed at the LLM-distillation reranking setting.

encode() and predict() also gained a per-call processing_kwargs override, so you can change processor settings like max_length, a vision-language model's image resolution, or a video's fps, for a single call without rebuilding the model.

The Agent Skill is the part of this release I'm most keen for people to try. Curious to hear how it works for you. I've been using it myself a lot to quickly set up some training runs that immediately use a bunch of best practices.

> pip install sentence-transformers==5.5.0
> hf skills add train-sentence-transformers

The full release notes: https://github.com/huggingface/sentence-transformers/releases/tag/v5.5.0

3 replies

·

danielhanchen

posted an update 2 months ago

Post

7784

We collaborated with NVIDIA to teach you how we made LLM training ~25% faster! 🚀

Learn how 3 optimizations help your home GPU train models faster:
1. Packed-sequence metadata caching
2. Double-buffered checkpoint reloads
3. Faster MoE routing

Guide: https://unsloth.ai/blog/nvidia-collab
GitHub: https://github.com/unslothai/unsloth

danielhanchen

posted an update 2 months ago

Post

8930

We made a guide on how to run open LLMs in Claude Code, Codex and OpenClaw.

Use Gemma 4 and Qwen3.6 GGUFs for local agentic coding on 24GB RAM

Run with self-healing tool calls, code execution, web search via the Unsloth API endpoint and llama.cpp

Guide: https://unsloth.ai/docs/basics/api

gg-tt

AI & ML interests

Recent Activity

Gemma 4 Technical Report

Gemma 4 Technical Report

Composer 2 Technical Report

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

Targeted Neuron Modulation via Contrastive Pair Search

Long Context Pre-Training with Lighthouse Attention

Efficient Pre-Training with Token Superposition

AI & ML interests

Recent Activity

Team members 95

gg-tt's activity